Terms of use
Plain terms for the site, the public API and the MCP server. Using any of them means accepting these.
What this is
OpenChainBench publishes reproducible benchmarks of crypto infrastructure: RPC providers, bridges, aggregators, price feeds, perpetual venues and prediction markets. Every number is produced by open-source code from a published methodology, and every benchmark says what it measured, when, and from where.
Licence
Benchmark data is published under CC-BY-4.0: reuse it anywhere, including commercially, as long as you credit OpenChainBench and link back to the benchmark page you took it from. The harness and site code are MIT licensed in the repository.
Not financial advice
Everything here is measurement, not recommendation. Nothing on this site, in the API or returned by the MCP server is investment, trading, legal or tax advice, an offer, or a solicitation to buy or sell any asset. We do not execute transactions, custody funds, route orders or act on your behalf, and we never will through these interfaces. A provider ranking first on a benchmark is a statement about one measured property under one methodology, not an endorsement.
Never send us an API key, a private key, a seed phrase or a signed transaction. No feature of this site requires one, and any request for one claiming to be from us is fraudulent.
Fair use of the API and MCP server
The public API and the MCP endpoint at https://openchainbench.com/api/mcp/mcp are read-only, free and need no account. They are rate limited per client so one caller cannot deny the service to others. Please do not attempt to bypass those limits, scrape at a rate that degrades the service for others, or present the data as your own measurements. Caching results on your side is encouraged.
Accuracy, and what we promise
Benchmarks are measured from a finite number of vantage points over a stated window. They can be affected by upstream outages, provider changes, network conditions and our own bugs. Figures are provided “as is”, without warranty of any kind, and we accept no liability for decisions taken on them. Each page states its window and sample size so you can judge the weight to give it; a benchmark whose data is stale is marked as such rather than silently served.
Corrections are public. If a number is wrong, open an issue in the repository or write to contact@openchainbench.com; the fix, and the reason for it, land in the commit history.
Disclosure
OpenChainBench is community-run. Mobula is measured here alongside its direct competitors under identical methodology. This disclosure is repeated on every relevant benchmark page and on /partners.
Changes
These terms change in public commits like everything else on the site. Material changes will be noted on this page. See also the privacy policy.